SUMMARY Non-small cell lung cancer (NSCLC) is an aggressive malignancy in which limited treatment options are further compromised by treatment resistance. Immunotherapy, particularly against the PD-1/PD-L1 immune checkpoint, has transformed NSCLC treatment with durable responses and comparatively minimal side effects in both second-line treatment of metastatic disease and, recently, first line therapy. Despite these impressive responses there are equally impressive but poorly defined intrinsic and/or acquired resistance mechanisms. Across solid tumor types, the response rate targeting the PD-1 axis in unselected patients is only ~20-30%. In previously untreated NSCLC, the overall response rate (ORR) to the anti-PD-1 antibody pembrolizumab is only ~45% even with patient pre-selection for >50% IHC PD-L1 tumor positivity, PD-L1 negative patients also exhibit anti-tumor response and the active search for alternative biomarkers of response in NSCLC has been unfulfilled. Thus, the significant resistance mechanisms impairing response to PD-1-targeted agents in NSCLC and other diverse solid tumors have remained intractable to both biomarker discovery and accompanying mechanistic definition. Project 2 of this U54 application thus directly addresses the pressing issue of intrinsic and acquired resistance to PD-1-targeted immunotherapy in NSCLC through analysis of an invaluable cohort of on-treatment longitudinal biopsies. Aim 1 pursues deep single-cell RNA-seq profiling of the immune component of NSCLC anti-PD-1 on-treatment biopsies using a highly efficient, microfluidic bead-based protocol allowing unsupervised discovery of cell clusters, T cell activation or exhaustion states and transcriptomic insights into immunotherapy resistance. Aim 2 exploits our 3D Patient-Derived Tumor Organoid (PDO) cultures that represent the first in vitro functional recapitulation of the PD-1-dependent immune checkpoint and tumor infiltrating lymphocytes (TILs) within clinical NSCLC biopsies. Here, we create functional organoid culture models of NSCLC immunotherapy resistance from longitudinal biopsies, measuring TIL activation upon in vitro anti-PD-1 organoid treatment and correlating against patient response. Lastly, Aim 3 performs prospective liquid biopsy and exome sequencing to determine mutational signatures of anti-PD-1 resistance that functionally regulate immune checkpoints. These Aims utilize synergistic expertise from the Stanford site of the U54 with Calvin Kuo (Project 2 PI; organoid culture, single cell RNA-seq), Ron Levy (tumor immunotherapy) and Heather Wakelee and Suki Padda (NSCLC immunotherapy trials), all in close coordination with Project 1 clinical biopsies, liquid biopsies and whole exome sequencing from Trever Bivona and Sourav Bandyopadhyay of the UCSF site. Overall, we present a comprehensive approach to intrinsic and acquired resistance to PD-1 inhibition in NSCLC via complementary single cell, organoid and sequencing analysis of longitudinal on-treatment biopsies.